One Place to Connect

Find Hold Without Taking, follow Jessica’s latest thinking, or send a note—all from one easy-to-use page.

A woman wearing a T-shirt featuring a parrot sits at an outdoor table with a white cockatoo perched on her shoulder.

Links, etc.

My First Novel Is Out, and It Started With Mushrooms I Didn’t Mean to Take

My First Novel Is Out, and It Started With Mushrooms I Didn’t Mean to Take

My first novel, Hold Without Taking, is out. It began with psilocybin I ordered by accident, a question about what a monkey puzzle tree might remember 800 years from now, and a boy named Kevin who had Duchenne muscular dystrophy. Here’s the honest story, including how I made it with BookyAI.

Is Bigger Better? Thirty-Seven AI Novels Say No

Is Bigger Better? Thirty-Seven AI Novels Say No

I generated the same 90,000-word novel 37 times across eight local AI models and three graphics cards, to find out which combination actually works on hardware you already own. Both halves of “bigger is better” turned out to be wrong — and the biggest improvement I found came from making a number smaller, not from buying anything.

Is AI Turning Books Into Open Source?

Is AI Turning Books Into Open Source?

I’m writing this from my raised-bed garden in West Lafayette, thinking about AI, copyright, and what happens to book publishing when the cost of writing a novel approaches zero. Are AI-generated books going the way of open-source software? Will marketing matter more than authorship? And can AI actually make creative opportunities more accessible? These are the questions behind my book-writing experiments, and I’d love for you to challenge my thinking.

What The Overstory Gets Wrong About Trees (And Why I Still Love It)

What The Overstory Gets Wrong About Trees (And Why I Still Love It)

Richard Powers’ The Overstory changed how I think about trees, fungi, memory, and the stories humans project onto nature. It also irritated me enough to help inspire Hold Without Taking, my speculative novel about a forest archive that remembers without understanding.

Two Models, Three GPUs: The Model Mattered More Than the Machine

Two Models, Three GPUs: The Model Mattered More Than the Machine

What matters more when running AI locally: the model or the machine? I tested two 8B models across three very different GPUs, from an RTX 4070 down to an RTX 2060, and the results surprised me. Faster hardware changed how long I waited. The model itself had a much bigger effect on what came out.

Let’s Connect

13 + 13 =